386 
PACIFIC SCIENCE, Vol. XVIII, October 1964 
tions ( Christopherson and Caum, 1931:20-39) 
includes only one malvaceous species, Sida fallax. 
Nor was tomentosum found on the small island 
of Molokini, which lies in the channel between 
Kahoolawe and Maui, by Forbes ( 1913<^: 10) . 
Collections from five of the eight major is- 
lands are preserved in the herbarium of the 
Bernice P. Bishop Museum. A total of 40 acces- 
sions includes no specimens from the other three 
islands, Niihau, Kauai, and Hawaii. Niihau is a 
privately owned cattle ranch, and few botanists 
have had the opportunity to visit it. Mrs. Sinclair 
(1885:23) included a painting of tomentosum 
in her "Indigenous Flowers of the Hawaiian 
Islands” which, according to St. John (1954: 
144-145 ) , was probably made from living ma- 
terial near her home in Keikei, Niihau. The only 
published record of tomentosum on Niihau is 
that of Forbes (1913^:23). 
I have been able to find only one definite 
record of the occurrence of this species on Kauai. 
This is a very old record (Wawra, 1873) which 
was kindly drawn to my attention by Dr. O. 
Degener. In recent years it has apparently not 
been seen, though St. John (loc. cit.) listed it 
among plants common to Niihau and Kauai. 
Ripperton and Hosaka (1942) did not record 
it from Kauai, and Mr. Stephen Au and Mr. 
A. W. Duvel, who are familiar with the botany 
of the island, inform me that they have never 
seen it there. Dr. C. M. Rick recently searched 
for it among the dry western foothills in the 
Mana-Kekaha region without success. This sug- 
gests that tomentosum is very rare, or possibly 
extinct, on Kauai. 
Concerning Hawaii, there is a rather puzzling 
conflict of evidence. MacCaughey (1917:414- 
418) described tomentosum as one of the species 
adapted to the colonization of lava flows, with 
an altitudinal range of 0-2000 ft. Although he 
did not state explicitly that the species occurred 
on Hawaii, his paper was concerned entirely 
with the ( geologically ) recent flows which are 
confined to Hawaii and East Maui. I have been 
able to find no independent record of the occur- 
rence of tomentosum in either of these areas. 
Further, an altitude of 2000 ft is a considerably 
higher elevation than any recorded for the 
Bishop Museum collections, or found during the 
present study. As will be evident later, tomento- 
sum is usually found as a component of Prosopis 
scrub, the altitudinal ranges of which have been 
mapped for the Hawaiian Islands by Ripperton 
and Hosaka (1942: maps 1-3). According to 
their maps there is one — and only one — region 
where this type of vegetation ("Zone A” in 
their terminology) extends to an elevation of 
2000 ft. This is the Kau Desert region in south- 
east Hawaii, and if MacCaughey’s statements 
concerning tomentosum applied to Hawaii, this 
would seem to be the only region which could 
have provided the necessary range in altitude. 
On the other hand, botanists who are very 
familiar with the Hawaiian flora (Miss Marie C. 
Neal, Mr. L. W. Bryan, and Mrs. C. K. Went- 
worth) have informed me that they have never 
seen the species on Hawaii, except in culture. It 
was not listed by Ripperton and Hosaka (1942) 
as a component of their "Zone A” vegetation on 
Hawaii, nor by Fagerlund and Mitchell (1944) 
in their check list of plants in the Hawaii 
National Park. 
The map in Figure 1 shows the probable 
geographic range of tomentosum as it exists 
today. Sites collected during the present study 
are indicated by solid circles. These do not 
differ materially from earlier collections recorded 
in the herbarium of the Bishop Museum. Sites 
recorded by earlier collectors, but not checked 
during the present study, are represented by 
solid circles enclosed in rings. The open circle 
shows the locations of hybrid populations on 
Oahu which will be considered later. Shaded 
portions of the map show the approximate areas 
in which the average annual rainfall is 20 inches 
or less (Leopold, 1951:2). 
In the case of the Oahu populations the ap- 
parent negative relation between rainfall and 
collection sites is quite significant, since the 
time available allowed for a careful search to be 
made on an island-wide basis, and no popula- 
tions were found outside these dry areas. Else- 
where, owing to time limitations, only the drier 
areas were carefully studied. However, none of 
the collections in the Bishop Museum for which 
locations are recorded, appears to have been 
collected outside the low rainfall areas, and it 
seems likely that in the neighbor islands, as in 
Oahu, rainfall is a primary factor limiting the 
range of the species. 
The negative relationship between rainfall 
and habitat is also indicated by the altitudinal 
